APPLICANT'S ABSTRACT: Brief interventions are needed to address the compounded problem of early sexual activity and alcohol use which results from the unsupervised social activities of young adolescents. Brief interventions which can reach a large number of adolescents with salient messages and/or increase the guidance by care givers, such as health providers, may decrease rates of unsupervised social drinking among adolescents. We aim to: 1) Revise our audio self-assessment and education program, which increased adolescent-provider discussion about condoms, increased condom use, and decreased signs of sexually transmitted diseases, so that it specifically addresses the social context of alcohol/HIV risk, 2) Conduct a randomized trial to determine whether a) an audio intervention alone or b) an audio + physician intervention reduces adolescent alcohol use and sexual behavior. Six primary care practices in Washington, D.C. health maintenance organizations (approximately 20 physicians) will participate in this study. Research staff will recruit approximately 18 of each physician's patients age 12 through 15 years. Adolescents will be randomized into three study groups: I-Usual Care, II-Audio Only, III-Audio + Physician. Physician-patient discussion about avoiding alcohol use and sex will be assessed by tap recordings of the general health examination and adolescent exit interviews. Adolescents' alcohol behaviors will be measured by telephone interview 4-, 8- and 12-months after the index physician visit. This study will demonstrate whether audio self-assessment alone or in conjunction with provider advice can decrease adolescent alcohol use and unprotected sexual intercourse. Although the study targets adolescents in primary care, the brief interventions could be more broadly administered if shown effective.